Next Article in Journal
Fabrication of Sustainable Sodium Alginate/Polyethyleneimine/Polyvinyl Alcohol Multilayer Composite Electrospun Nanofiber Membrane for Efficient Cu2+ Removal
Previous Article in Journal
Exploring the Competencies, Phases and Dimensions of Municipal Administrative Management towards Sustainability: A Systematic Review
Previous Article in Special Issue
Travel Mode Choice Prediction to Pursue Sustainable Transportation and Enhance Health Parameters Using R
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Analyzing the Impact of Montreal’s Réseau Express Vélo (REV) on Surrounding Bike Lanes’ Ridership and the COVID-19 Cycling Recovery

1
Department of Urban Environments, Université de l’Ontario Français, 9 Lower Jarvis St., Toronto, ON M5E 1Z2, Canada
2
Department of Urban Studies and Tourism, Université du Québec à Montréal, Montreal, QC H3C 3P8, Canada
*
Author to whom correspondence should be addressed.
Sustainability 2024, 16(14), 5992; https://doi.org/10.3390/su16145992
Submission received: 29 May 2024 / Revised: 8 July 2024 / Accepted: 11 July 2024 / Published: 13 July 2024
(This article belongs to the Collection Sustainable Urban Mobility Project)

Abstract

:
Amidst the COVID-19 pandemic, Montreal implemented pro-cycling measures and enhanced its cycling infrastructure, notably by the introduction of the Réseau Express Vélo (REV), an extensive network of separated high-capacity bike lanes inaugurated in 2020. This paper delves into the pandemic’s impact on Montreal’s cycling network, with specific focus on the Berri/Lajeunesse/Saint-Denis REV route, evaluating its influence on the recovery of adjacent bike lanes and its effectiveness in attracting new cyclists. Using data from sensors installed along Montreal’s bike lanes between 2018 and 2023, our analysis reveals an initial average 28% decline in cycling volumes at the onset of the pandemic. However, from 2021 onwards, disparities began to emerge. While sensors on adjacent parallel bike routes to REV recorded further declines in ridership, those elsewhere in the city showed signs of cycling recovery, suggesting a shift in the cycling patterns towards the REV. Distinguishing between cyclists displaced from other parallel bike lanes and those representing a pent-up demand, our study indicates that the displaced cyclists accounted for 64% of the REV’s ridership at the southernmost sensors and only 7% at the northernmost sensors in 2023. These findings have significant policy implications, as cyclists comprised within the pent-up demand category correspond to those having transitioned from not using bike lanes to utilizing safer cycling infrastructure and account for the net growth in the cycling mode share directly attributable to the REV. Both of these observations are pivotal in fostering a shift away from cars and steering Montreal towards achieving its cycling mode share goal of 15% by 2027.

1. Introduction

In an effort to reduce vehicle emissions and alleviate traffic congestion, many cities have turned their sights towards pro-bicycle policies and cycling infrastructures. In hopes of increasing the share of cyclists, cities have developed and progressively expanded extensive networks of bicycle lanes to enhance the safety and travel experience of cyclists. The pandemic and observed surge in leisure cycling has further prompted some cities to seize this moment as an opportunity to expedite the overhaul of existing cycling networks and increase the number of protected bike lanes in their regions.
Montreal was no exception to this trend, as it used the backdrop of the pandemic to implement pro-cycling measures, such as bidirectional protected pop-up bike lanes installed in 2020 and the introduction of the planned express bikeway network titled the Réseau Express Vélo (REV). Inaugurated in November 2020, the REV is a network of separated high-capacity bike lanes opened year-round. With over 20 km already completed, this 191 km project offers a safe, efficient, and pleasant cycling experience, which is expected to help the city achieve its goal of increasing the cycling mode share to 15% of all trips within its central boroughs by 2027 [1].
The primary axis in the current REV network is the Berri/Lajeunesse/Saint-Denis route, situated along one of Montreal’s most renowned commercial streets. While studies have examined its impact on the local economy and tallied cyclists along this route [2,3], little attention has been devoted to understanding its effect on the adjacent cycling network.
Using data from sensors installed along Montreal’s bike lanes between 2018 and 2023, we examine the impact the Saint-Denis REV has had on the surrounding bike lanes. We dissect the yearly cyclist count along this route to uncover the extent at which this bike lane has been able to generate new cyclists and the proportion of cyclists captured from adjacent parallel bike routes. Having data that span the periods before and during the COVID-19 pandemic, we also explore the effect of the pandemic on Montreal’s cycling network and investigate whether routes parallel to the REV recovered at a different rate.
In the following section, we examine the literature on protected bike lanes and its implications for cycling uptake both on the route itself and on the remainder of the cycling network. The impact of the COVID-19 pandemic on the cycling mode share is also considered. Section 3 presents the history and context of Montreal’s REV express bikeway network and describes our data and methodology for assessing its impact on adjacent bike lanes. The effect of the Saint-Denis REV on the surrounding parallel bike lanes as well as the proportion of its users captured from adjacent bike lanes is presented in Section 4. These findings are accompanied by the results of our investigation on whether the REV caused the nearby parallel bike lanes to recover from the pandemic at a different rate. We discuss these findings in Section 5, focusing primarily on their implications for cities.

2. Literature Review

Numerous studies show that adding cycling infrastructures can significantly increase the number of cyclists [4,5,6,7,8]. This increase is observed not only on the new facilities but also on the network in general. Commonly referred to as the network effect, the increase in cyclists along the new cycling route will often translate to higher ridership along adjacent connected bike lanes as well [5,9,10].
The interconnected nature of the network can also prompt cyclists to alter their routes, especially when a new infrastructure offers greater efficiency or safety compared to the existing routes. It is therefore implicitly assumed that a portion of the ridership on new routes consists of diverted existing cyclists. As explained by De Jong, Böcker, and Weber [11], cyclists choose their route based on the available infrastructures (e.g., separated lanes) and the routes’ surroundings (e.g., green areas). De Jong, Böcker, and Weber [11] found that these attributes can lead cyclists to deviate from their usual route and, on average, cover 59% more distance than that corresponding to the shortest path. In the same vein, Dill and Gliebe [12] showed that, compared to the shortest route, the median additional distance cyclists are willing to cover is equal to 0.437 km. In addition, they observed a clear preference for separated or protected lanes. Larsen and El-Geneidy [13] indeed found such a preference and showed that, compared to the distance covered on unprotected bike lanes, cyclists would on average travel 1 km further or 16% longer to reach a protected bike lane. In another study, Krizek, El-Geneidy, and Thompson [14] found that the willingness to deviate would lead on average to a 67% longer distance to reach an off-street separated lane. In terms of time, Tilahun et al. [15] showed that cyclists would travel up to 20 min more in order to reach such forms of protected cycling infrastructure. These results suggest that some cyclists may alter their current route to take advantage of a safer cycling infrastructure, and the extent to which this is true will be examined in this paper using cycling count data from the REV and adjacent parallel bike routes in Montreal.
Research indicates that cyclists exhibit a greater propensity to deviate from their usual route and opt for a longer yet safer path when undertaking longer trips and in areas where the occurrence of accidents is more prevalent [12]. The sense of safety is crucial, as several studies show that certain individuals (e.g., inexperienced cyclists) express fear or reluctance towards cycling near motorized vehicles. These individuals would often not cycle at the same extent if safe cycling infrastructures were not present [16,17,18]. Hence, the higher capacity of attraction of separated bike lanes is not surprising, as they have been shown to reduce the number of accidents and injuries while enhancing a sense of safety [19,20,21]. By enabling more people to cycle, such bike infrastructures foster a virtuous cycle whereby the increased presence of cyclists on a bike route heightens driver awareness and encourages more a cautious behavior around them [22,23].
The provision of protected cycle paths is an important contributor in the assessment of bikeability, a concept that measures how conducive a given environment is to bicycling [24,25,26,27]. As discussed by Castañon and Ribeiro [25], measuring bikeability typically involves using indicators and indexes to estimate attributes such connectivity, accessibility, convenience, and safety [25,28]. To our knowledge, no studies have yet examined the bikeability of the REV. We hope that examining its rider composition and impact on adjacent bike lanes will provide a foundational basis for further research in this area.
Many cities around the world are adopting sustainable mobility strategies that aim to implement and promote an optimal mix of transportation modes to maximize the collective well-being in environmental, social, and economic terms [29]. Cycle paths generate numerous benefits, notably by alleviating congestion and reducing polluting emissions while improving the health of the citizens. They therefore represent efficient infrastructure investments compared to road transport, forming an integral part of a sustainable mobility strategy [30]. Under this interpretation of sustainability, transport policies will contribute to achieving strong sustainable development if they generate social and economic benefits while preserving the natural environment from negative impacts [31,32]. Given the efficiency of cycling infrastructure investments in terms of sustainability, many cities decided to expand their cycling network and promoted cycling during the pandemic [33,34]. Washington, D.C., for instance, managed to build a bike network of approximately 150 km from 2001 to 2019 but added another 30 km of protected bike lanes between 2020 and 2021 alone [35]. Paris was also able to bypass the standard construction process and rapidly expand its cycling network during the COVID-19 pandemic. At the onset of the pandemic, it installed 52 km of temporary bike lanes separated from traffic, which were then subsequently made permanent [36]. From 2020 to 2021, the city of Austin, Texas, built 47 km of separated protected bike lanes, while Barcelona removed 21 km of car lanes replacing them with pop-up bike lanes [37].
In general, studies conducted during the pandemic period have revealed that the number of cyclists increased on weekends but not necessarily during the week due to reduced travel associated with work and school closures [33,38,39]. However, the impact of the pandemic has not been consistent over time. For instance, a recent study examining the effect of the pandemic on the cycling volumes in Montreal reported a 2% increase in cycling volumes between 2019 and 2020, followed by an 8% decline from 2020 to 2021 [37]. However, scant attention has been paid to investigating the pandemic’s effect on cycling at a more granular level, particularly regarding the role of newly implemented cycling infrastructure in improving cycling recovery. In this paper, we aimed to explore the impact of the COVID-19 pandemic on Montreal’s cycling network and assess whether routes parallel to the REV recovered at a different rate. This pandemic-linked recovery was further dissected to establish the proportion of the REV’s ridership likely captured from the surrounding bike lanes and the REV’s ability to generate new users.

3. Methodology

3.1. Case Study Area

This study took place in the city of Montreal, which is the second largest city in Canada, boasting a metropolitan population of over 4.2 million residents [40]. Montreal has long been ranked among North America’s top cycling cities [41] and stands out with the highest cycling mode share among major Canadian cities, with cycling accounting for 3.3% of all commuting trips in the city [42]. Furthermore, 36.4% of residents aged 18–74 years reported cycling at least once a week, as per the latest study conducted by the Autorité régionale de transport métropolitain [43], and over 10% of them cycle year-round, showcasing a commitment to cycling despite the city’s winter conditions. Montreal’s existing cycling network comprises over 900 km of bike infrastructures, featuring 218 km of protected bike lanes, a network of bike routes where cyclists are given priority over cars, as is the case for the REV express bikeway network. With the first stage opened in November 2019, the 191 km REV protected bike network aims to provide an efficient and safe mean of cycling within the city and foster a notable modal shift away from cars. To accomplish this, it provides wide lanes that facilitate overtaking and specific traffic signals that prioritize cyclists over cars and utilizes concrete medians to separate cyclists from vehicular traffic.
The comfort provided by the REV is further believed to attract potential cyclists who have not yet adopted this mode of travel and help achieve the city’s objective of a 15% cycling modal share by 2027 [1]. Since opening in November 2020, the REV network has emerged as a success, with each of its routes seeing significant ridership levels, and the primary 8.7 km axis along Saint-Denis Street even breaking longstanding ridership records for existing bike routes, surpassing 1.5 million trips in 2023 [44]. However, while impressive, these ridership figures offer only a preliminary insight into how the REV has influenced the cycling rates within the city and do not account for its impact on the surrounding cycling infrastructures. To better understand the influence that the REV has had on the Montreal cycling network, and particularly on its most heavily used segment, we examined its effects on adjacent parallel bike routes. We show its ability to attract potential cyclists by uncovering the proportion of its ridership likely captured from the surrounding bike lanes. We further examined the impact of the COVID-19 pandemic on Montreal’s cycling network and investigate whether routes parallel to the REV recovered at a different rate.

3.2. Data and Methods of Analysis

The data for this study were collected from sensors installed along bike lanes in Montreal from 2018 to 2023. This dataset is available on the City of Montreal’s open data portal and provides the hourly cycling ridership levels recorded by each sensor [45]. The sensors themselves were installed by the Eco-Counter company that supplies sensors to the city, most of which are placed underground, and some also physically display the data alongside the counter. Previous research using these data revealed variations in cycling ridership [37] and highlighted the growth in ridership over time along the Saint-Denis REV corridor [3]. However, these investigations faced challenges due to missing or, at times, inaccurate data, issues well documented on the City of Montreal’s open data portal. To overcome this challenge, we identified and removed problematic data entries, which were deemed faulty beyond repair and consisted primarily of post-COVID-19 data that could not be imputed due to a lack of prior observations or sensor data identified as unreliable by Eco-Counter. When possible, we imputed the missing values using the na_seasplit function in the imputeTS R package [46]. This approach to missing value imputation, referred to as seasonally splitted missing value imputation, involves segmenting time series into seasonal intervals and subsequently applying linear interpolations independently to impute the missing values within each defined segment. We subsequently removed the sensors flagged by Eco-Counter as inaccurate or that possessed excessively large data gaps rendering the imputations inaccurate. Following this methodology, we retained a total of 21 sensors with accurate data covering our study period.
To understand the impact of the Saint-Denis REV on the ridership of adjacent parallel bike routes and to establish the extent to which it has attracted new cyclists, we first examined the effect of the COVID-19 pandemic on Montreal’s cycling network by considering each sensor’s ridership over time. We were particularly interested in establishing whether the sensors positioned along parallel bike routes adjacent to the REV recorded a different rate of recovery compared to those placed elsewhere. Our hypothesis was that recovery on parallel routes adjacent to the REV took longer—or is still ongoing—in comparison to the average recovery observed for the rest of the cycling network. To conduct this analysis, the recovery rates measured by 6 sensors situated along bike routes parallel to the REV were compared to those recorded by 11 control sensors dispersed throughout the city. Other than the abovementioned four REV sensors and six adjacent parallel sensors, all sensors identified by Eco-Counter as reliable with complete data from 2018 to 2022 were included within the control group. The location of these sensors, as well as the four situated along the REV, is displayed in Figure 1.
To better understand the REV’s effect on adjacent bike lanes and shed light on its potential to generate new users, we also examined its ridership in more detail. To establish the proportion of its users that are newly generated or displaced from the surrounding parallel bike routes, we first calculated the expected drop in cycling ridership at the onset of the COVID-19 pandemic on the surrounding parallel routes. The expected drop in cycling ridership associated with the pandemic was determined by averaging the drop observed on the Montreal cycling network, excluding that measured by the sensors located along the Saint-Denis REV or nearby parallel bike routes. More specifically, the approach consisted of comparing the ridership level indicated by the sensors for each year following the pandemic (2020 to 2023) to the average of the two years prior the pandemic (2018–2019). We thus obtained the percentage change in ridership per year following the pandemic, which, when averaged, could tell us what we could expect to see on bike routes near the REV were it not to exist.
The difference between the observed and the expected drop in cycling ridership along parallel routes corresponds to the number of cyclists transferred from nearby parallel bike route towards the REV. We refer to this group as displaced cyclists, and by subtracting this number from its corresponding REV sensor ridership count, we obtained the pent-up demand for safe cycling infrastructures. This pent-up demand category consists of cyclists captured from the surrounding streets not comprised in Montreal’s cycling network as well as new cyclists who began to bike because of the REV. These two groups of cyclists are particularly important from a policy perspective, as the former represents REV users who transitioned from not using a bike lane to using a safer cycling infrastructure, and the latter accounts for the net growth in the cycling mode share directly attributable to the REV.
The methodology to calculate displaced cyclists and pent-up demand is depicted in Figure 2. By subtracting the observed ridership in the current period from the average ridership during the pre-COVID-19 period, we obtained the drop in actual ridership, shown in red. We could then compare this drop with the expected loss in ridership due to the COVID-19 pandemic, illustrated in purple, that would be measured if the sensors’ count declined at the same rate as that of the average ridership measured by the control sensors. The difference between the actual drop in cycling ridership and the loss expected due to the pandemic is referred to as displaced cyclists, depicted in yellow. By subtracting this group of cyclists from the corresponding REV sensors’ ridership count, we obtained the pent-up demand for safe cycling infrastructures, illustrated in green.
Fortunately, each parallel sensor is situated near a corresponding sensor along the REV. This allowed for a more direct comparison and a more accurate assessment of the displaced cyclist counts. To facilitate the interpretation of the results, we grouped the parallel sensors and their corresponding REV sensors into four clusters, as illustrated in Figure 1. The analysis of displaced cyclists and pent-up demand was conducted for each of these four clusters to provide a more detailed understanding of the REV’s impact on Montreal’s cycling network and the overall cycling mode share.

4. Results

4.1. REV’s Effects on Montreal’s Cycling Network Recovery from the COVID-19 Pandemic

Table 1 shows the ridership level measured by all sensors included in our analyses (N = 21). The greyed-out boxes represent time periods for which the sensor data were unavailable or, in the case of the REV, periods predating the installation of the sensors. Since its inauguration in 2021, the Berri/Lajeunesse/Saint-Denis REV route has witnessed a high cyclist demand and has become the most used cycling route in the city, surpassing 1.5 million rides at the level of Des Carrières Avenue (Cluster B) according to the 2023 sensor readings. Sensors situated along the REV in Clusters A and B appeared to record much higher ridership levels than those located in Clusters C and D, which we attributed to their proximity to downtown and other major cycling routes nearby.
Table 2 shows the percent change in cycling ridership from the pre-COVID-19 period. This was calculated by subtracting the yearly ridership level from that of the average ridership of 2018–2019. Each presented percentage is a direct comparison to that of the average pre-COVID-19 ridership level. By weighing each sensor’s percentage change by its respective ridership level, we found a fairly equivalent drop in ridership for both the control sensors (−28.7%) and the parallel sensors (−28.3%) in 2020 following the pandemic and ensuing lockdown measures. What is notable is that the control counts, on average, tended to return to the pre-pandemic ridership levels at a much faster rate than those determined by the sensors located on bike routes parallel to the REV. In fact, while the loss in ridership along the routes where the control sensors were installed appeared to recede every year, the opposite trend was observed along parallel routes. On average, the levels measured by the control sensors increased by 3.4% compared to the pre-COVID-19 ridership levels in 2023, whereas the ridership levels on the parallel routes for that same year remained 35% lower than the pre-COVID-19 levels. If anything, the parallel routes seemed to be losing more cyclists as time progressed.
To further investigate the relation between parallel and control routes’ ability to recover from the pandemic, we conducted a Pearson correlation test. The results revealed a statistically significant negative correlation between the yearly increase in ridership since the pandemic along the control bike routes and the continual loss in ridership along the parallel bike routes (r = −0.96, p < 0.036). This further indicated that Montreal’s cycling network as a whole is well on its way to returning to the pre-pandemic ridership levels, but the bike routes parallel to the REV may be experiencing a decline in ridership as cyclists are drawn towards the REV. The following section delves deeper into the REV’s ridership to determine the proportion of cyclists likely drawn from the surrounding bike lanes and its potential to attract new users.

4.2. REV Ridership Analysis

It only took three years to the Berri/Lajeunesse/Saint-Denis REV route to become the most popular cycling artery in the city. However, asserting that it generated new riders and increased Montreal’s overall cycling mode share by simply counting its 1.5 million riders would be erroneous, as a portion of these cyclists were drawn from the surrounding bike routes.
To uncover the true effect of the REV, we examined the loss in ridership on the surrounding parallel bike routes after accounting for the impact of the pandemic. The expected drop in ridership due to COVID-19 was calculated by averaging the drop in ridership indicated by the control sensors for each given year, which, when subtracted from the actual drop in ridership along the parallel routes gave us the drop in ridership associated with the REV. In other words, we obtained the number of cyclists who switched from parallel bike routes to the REV, which we, again, refer to as displaced cyclists. To ensure an accurate comparison, we doubled the ridership count of the SaintUrbain_Villeneuve parallel route to adjust for its single directional (south) nature, unlike the REV, which is bidirectional. The yearly expected drop in ridership and the actual drop in ridership along the parallel routes per cluster is depicted in Table 3.
As shown in Table 3, the observed drop in ridership on the parallel routes consistently exceeded the expected drop attributed to the pandemic. This suggests that some cyclists changed their preferred route following the implementation of the Berri/Lajeunesse/Saint-Denis REV route and that a portion of the REV’s ridership thus comprised existing cyclists having simply adjusted their route choice, rather than entirely new riders.
By subtracting the number of displaced cyclists from the overall ridership measured by each corresponding REV sensor, we obtained the pent-up demand. This category encompasses both cyclists redirected from non-protected bike routes to the REV as well as new cyclists generated by the REV. These users are particularly important to policymakers, as they now have access to safer cycling infrastructure and directly contribute to increasing the city’s cycling mode share. Figure 3 provides a breakdown of the distribution of displaced cyclists and pent-up demand over time for each of the four REV sensors.
As shown in Figure 3, the sensors positioned along the REV in Clusters A and B exhibited significantly higher ridership compared to those in Clusters C and D, which we attributed to their proximity to downtown and other major cycling routes. Clusters A and B also showed a higher share of displaced cyclists, but when looking at the number of cyclists comprised within the pent-up demand, we found that the REV sensors located in Clusters B and C generated most of the pent-up demand (accounting for over 100,000 counts during some months in 2023 and 2021 in the case of Cluster B). More details on the number and proportion of cyclists comprised within the displaced and pent-up demand categories per REV sensor over time are provided in Table 4.
As can be seen in Table 4, the number of displaced cyclists per sensor seemed to increase over time as the parallel routes continually struggled to regain their pre-COVID-19 ridership levels. Meanwhile, the pent-up demand appeared to rise for Clusters A and C and remain relatively constant for the other two clusters throughout the study period, which suggests that the REV managed to attract new users, but more importantly, it also succeeded in retaining them. That said, because the growth in displaced cyclists exceeded the increase in pent-up demand, the proportion of the REV’s ridership attributed to the pent-up demand appeared to recede as time progressed.
What is evident from this analysis is the increasing popularity of the Berri/Lajeunesse/Saint-Denis REV route, as it has continued to capture cyclists from the adjacent parallel routes. The REV’s ridership includes both existing cyclists having altered their choice of route and new riders; by discerning the proportion of total ridership attributed to each group, we now have a better understanding of its potential impact on the city’s cycling infrastructure and overall cycling mode share.

5. Discussion and Concluding Remarks

While the impacts of the COVID-19 pandemic on urban cycling continue to manifest, there does seem to be a consensus that leisure cycling soared during the earlier months of the pandemic and that the cycling mode share has grown in cities that seized the pandemic as an opportunity to enhance their existing cycling network [37]. Nonetheless, studies underscored the fluctuating nature of these effects over time and highlighted how the observed increase in leisure cycling may have been offset by other pandemic-driven shifts in travel behavior, such as an uptick in remote work. In this paper, we examined how the pandemic affected Montreal’s cycling network, focusing specifically on the recently introduced Berri/Lajeunesse/Saint-Denis REV route to understand how this new cycling infrastructure has influenced the recovery of the adjacent parallel bike routes and assess its effectiveness in attracting new cyclists.
By analyzing the cycling count data collected from 21 sensors in Montreal, we found that the cycling volumes diminished at the onset of the COVID-19 pandemic and have not yet returned to their pre-pandemic levels. Specifically, the cycling ridership diminished by 28.5% in 2021 in comparison to the average of 2018–2019 and remained lower by 12.8% in 2023 than the baseline level of 2018–2019. Fortunately, there are signs that the reduction in the cycling volumes attributed to the pandemic is gradually diminishing, as indicated by the smaller percentage drops in overall ridership throughout our study illustrated in Table 5.
The overall declines in ridership observed in our analysis are notably higher than those reported by Buehler and Pucher [33,37]. In their study using data from Eastern Canada, primarily collected from sensors located in Montreal, they found a 10% decrease in 2020 compared to 2019, followed by a 3% increase between 2019 and 2021. We attribute this difference in results to us accounting for the introduction of additional sensors and for periods of missing data, as outlined in [47]. However, we recognize that the scope, timeliness, and context of earlier studies may not have allowed for this level of analysis. We hope that our research will offer a more accurate understanding of the cycling trends in Montreal, enabling the city to plan accordingly.
Upon closer examination of the changes in the cycling volumes, it becomes apparent that these changes were not uniform across all sensors, with some routes experiencing different levels of recovery than others. In 2020, at the onset of the pandemic, both control and parallel sensors recorded relatively similar declines in ridership. However, starting in 2021, disparities began to emerge. The sensors located on the adjacent parallel routes to the newly introduced Berri/Lajeunesse/Saint-Denis REV route continued to measure even greater declines in ridership, while the sensors situated elsewhere in the city began to show signs of ridership recovery. This trend persisted and became even more pronounced over the years, with the parallel routes experiencing a 35.5% decrease in cycling volume between 2018–2019 and 2023, while the control sensors registered a 1.9% increase during that same period. We attribute this difference to a portion of cyclists along the parallel routes now altering their behavior in favor of the REV.
The Berri/Lajeunesse/Saint-Denis REV route is now the most popular bike route in Montreal, but ascertaining its positive effect on the cycling mode share by simply counting its overall ridership is disingenuous, as a portion of these cyclists were drawn from the adjacent bike routes. This group is referred to as displaced cyclists and accounted for 64% of the REV’s ridership recorded by its southernmost sensor (Cluster A) and only 7% of that measured by its northernmost sensor (Cluster D) in 2023. By subtracting this number of cyclists from each REV’s sensors overall ridership, we obtained the pent-up demand for safe cycling infrastructures. This pent-up demand category encompasses cyclists drawn from the surrounding streets not included in Montreal’s cycling network as well as new cyclists who started biking because of the REV. These two groups of cyclists are particularly significant from a policy standpoint, as the former represents REV users who transitioned from not using a bike lane to utilizing safer cycling infrastructure, while the latter accounts for the net growth in the cycling mode share directly attributable to the REV.
The proportion of the REV’s ridership accounted for by the pent-up demand recorded by the two northernmost sensor clusters (C and D) was much higher, which may indicate that these areas were not well served by the existing cycling infrastructures and that this is where the REV has led to a more pronounced increase in the cycling mode share. The higher proportion of displaced cyclists determined by the two lower sensor clusters (A and B) is also noteworthy, as it suggests that a considerable share of cyclists were willing to alter their travel behaviors in order to take advantage of the wide cycling infrastructure fully protected from cars. The pent-up demand recorded by Clusters A and C increased, while that measured by Clusters B and D remained relatively constant, which suggests that the REV has not only succeeded in transitioning cyclists towards safer routes and attracting new users but also been effective in retaining them along this secure path. However, since the growth in displaced cyclists has consistently surpassed the rise in pent-up demand, the proportion of REV’s ridership attributed to the pent-up demand seems to have diminished over time. This suggests that cyclists are still discovering the advantages of the REV and adjusting their travel patterns accordingly. However, we anticipate this portion of the overall REV ridership to level off as soon as ridership along the adjacent parallel routes stabilizes.
The Berri/Lajeunesse/Saint-Denis REV is one of the first routes in the proposed 191 km REV protected bike network, aimed at providing an efficient and safe means of cycling within the city. As additional routes are added to the network, the current REV route is likely to benefit from network effects, as more cyclists will utilize the enhanced cycling network. An interesting venue for future research is to investigate whether the addition of other protected bike routes in the city will result in a higher pent-up demand along the Berri/Lajeunesse/Saint-Denis REV route and other protected bike routes within the REV network. This would help determine whether Montreal is being successful in fostering a notable modal shift away from cars and achieving the targeted 15% cycling mode share in central boroughs by 2027.

Author Contributions

Conceptualization, M.Y., G.M. and G.A.T.; methodology, M.Y., G.M. and G.A.T.; formal analysis, M.Y. and G.M.; data curation, G.M.; writing—original draft preparation, M.Y. and G.A.T.; writing—review and editing, M.Y. and G.A.T.; visualization, G.M.; supervision, M.Y. and G.A.T. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original data presented in the study are openly available at https://donnees.montreal.ca/dataset/reseau-express-velo (accessed on 7 December 2023). All the R codes used in this paper are available on the author’s GitHub repository at https://github.com/Gavin-MacG/Cycling_StDenisREV.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Ville De Montréal. The REV: An Expressed Bike Network. 2022. Available online: https://montreal.ca/en/articles/rev-express-bike-network-4666 (accessed on 12 December 2023).
  2. Freixo, Q. REV Saint-Denis: Boosting the Local Economy through Cycling and Walking in Montreal, QC; Eco-Compteur: Lannion, France, 2023; (accessed on 3 February 2024). [Google Scholar]
  3. Chapalain, R. Saint-Denis Street Redevelopment: Stimulating the Local Economy through Cycling and Walking in Montreal, Qc. Eco-Compteur. 2023. Available online: https://www.eco-counter.com/blog/saint-denis-street-stimulating-local-economy-cycling-walking-montreal/ (accessed on 3 May 2024).
  4. Mölenberg, F.J.; Panter, J.; Burdorf, A.; van Lenthe, F.J. A systematic review of the effect of infrastructural interventions to promote cycling: Strengthening causal inference from observational data. Int. J. Behav. Nutr. Phys. Act. 2019, 16, 93. [Google Scholar] [CrossRef] [PubMed]
  5. Buehler, R.; Dill, J. Bikeway networks: A review of effects on cycling. Transp. Rev. 2016, 36, 9–27. [Google Scholar] [CrossRef]
  6. Young, M.; Savan, B.; Manaugh, K.; Scott, J. Mapping the demand and potential for cycling in Toronto. Int. J. Sustain. Transp. 2021, 15, 285–293. [Google Scholar] [CrossRef]
  7. Félix, R.; Cambra, P.; Moura, F. Build it and give ‘em bikes, and they will come: The effects of cycling infrastructure and bike-sharing system in Lisbon. Case Stud. Transp. Policy 2020, 8, 672–682. [Google Scholar] [CrossRef]
  8. Buehler, R.; Pucher, J. Cycling to work in 90 large American cities: New evidence on the role of bike paths and lanes. Transportation 2012, 39, 409–432. [Google Scholar] [CrossRef]
  9. Lowry, M.; Loh, T.H. Quantifying bicycle network connectivity. Prev. Med. 2017, 95, 5134–5140. [Google Scholar] [CrossRef] [PubMed]
  10. Furth, P.G.; Mekuria, M.C.; Nixon, H. Network connectivity for low-stress bicycling. Transp. Res. Rec. 2016, 2587, 41–49. [Google Scholar] [CrossRef]
  11. De Jong, T.; Böcker, L.; Weber, C. Road infrastructures, spatial surroundings, and the demand and route choices for cycling: Evidence from a GPS-based mode detection study from Oslo, Norway. Environ. Plan. B Urban Anal. City Sci. 2023, 50, 2133–2150. [Google Scholar] [CrossRef]
  12. Dill, J.; Gliebe, J. Understanding and Measuring Bicycling Behavior: A Focus on Travel Time and Route Choice; Final Report OTREC-RR-08-03 Prepared for Oregon Transportation Research and Education Consortium (OTREC); OTREC: Portland, OR, USA, 2008. [Google Scholar] [CrossRef]
  13. Larsen, J.; El-Geneidy, A. A travel behavior analysis of urban cycling facilities in Montréal, Canada. Transp. Res. Part D Transp. Environ. 2011, 16, 172–177. [Google Scholar] [CrossRef]
  14. Krizek, K.J.; El-Geneidy, A.; Thompson, K. A detailed analysis of how an urban trail system affects cyclists’ travel. Transportation 2007, 34, 611–624. [Google Scholar] [CrossRef]
  15. Tilahun, N.Y.; Levinson, D.M.; Krizek, K.J. Trails, lanes, or traffic: Valuing bicycle facilities with an adaptive stated preference survey. Transp. Res. Part A Policy Pract. 2007, 41, 287–301. [Google Scholar] [CrossRef]
  16. Pearson, L.; Berkovic, D.; Reeder, S.; Gabbe, B.; Beck, B. Adults’ self-reported barriers and enablers to riding a bike for transport: A systematic review. Transp. Rev. 2023, 43, 356–384. [Google Scholar] [CrossRef]
  17. Heinen, E.; Van Wee, B.; Maat, K. Commuting by bicycle: An overview of the literature. Transp. Rev. 2010, 30, 59–96. [Google Scholar] [CrossRef]
  18. Garrard, J.; Rose, G.; Lo, S.K. Promoting transportation cycling for women: The role of bicycle infrastructure. Prev. Med. 2008, 46, 55–59. [Google Scholar] [CrossRef] [PubMed]
  19. Guirao, B.; Gálvez-Pérez, D.; Casado-Sanz, N. The impact of the cyclist infrastructure type on bike accidents: The experience of Madrid. Transp. Res. Procedia 2023, 71, 403–410. [Google Scholar] [CrossRef]
  20. Thomas, B.; DeRobertis, M. The safety of urban cycle tracks: A review of the literature. Accid. Anal. Prev. 2013, 52, 219–227. [Google Scholar] [CrossRef]
  21. Harris, M.A.; Reynolds, C.C.; Winters, M.; Cripton, P.A.; Shen, H.; Chipman, M.L.; Cusimano, M.D.; Babul, S.; Brubacher, J.R.; Friedman, S.M.; et al. Comparing the effects of infrastructure on bicycling injury at intersections and non-intersections using a case–crossover design. Inj. Prev. 2013, 19, 303–310. [Google Scholar] [CrossRef]
  22. Jacobsen, P.L. Safety in numbers: More walkers and bicyclists, safer walking and bicycling. Inj. Prev. 2015, 21, 271–275. [Google Scholar] [CrossRef] [PubMed]
  23. Nordback, K.; Marshall, W.E.; Janson, B.N. Bicyclist safety performance functions for a US city. Accid. Anal. Prev. 2014, 65, 114–122. [Google Scholar] [CrossRef] [PubMed]
  24. Wysling, L.; Purves, R.S. Where to improve cycling infrastructure? Assessing bicycle suitability and bikeability with open data in the city of Paris. Transp. Res. Interdiscip. Perspect. 2022, 15, 100648. [Google Scholar] [CrossRef]
  25. Castañon, U.N.; Ribeiro, P.J.G. Bikeability and Emerging Phenomena in Cycling: Exploratory Analysis and Review. Sustainability 2021, 13, 2394. [Google Scholar] [CrossRef]
  26. Krenn, P.; Oja, P.; Titze, S. Development of a Bikeability Index to Assess the Bicycle-Friendliness of Urban Environments. Open J. Civ. Eng. 2015, 5, 451–459. [Google Scholar] [CrossRef]
  27. Fosgerau, M.; Łukawska, M.; Paulsen, M.; Rasmussen, T.K. Bikeability and the induced demand for cycling. Proc. Natl. Acad. Sci. USA 2023, 120, e2220515120. [Google Scholar] [CrossRef] [PubMed]
  28. Codina, O.; Maciejewska, M.; Nadal, J.; Marquet, O. Built environment bikeability as a predictor of cycling frequency: Lessons from Barcelona. Transp. Res. Interdiscip. Perspect. 2022, 16, 100725. [Google Scholar] [CrossRef]
  29. Cartenì, A. Urban sustainable mobility. Part 1: Rationality in transport planning. Transp. Probl. 2014, 9, 39–48. [Google Scholar]
  30. Henke, I.; Molitierno, C.; Errico, A. Decision-Making in the Transport Sector: A Sustainable Evaluation Method for Road Infrastructure. Sustainability 2020, 12, 764. [Google Scholar] [CrossRef]
  31. Cartenì, A.; Henke, I.; Molitierno, C.; Di Francesco, L. Strong Sustainability in Public Transport Policies: An e-Mobility Bus Fleet Application in Sorrento Peninsula (Italy). Sustainability 2020, 12, 7033. [Google Scholar] [CrossRef]
  32. Shah, S.A.R.; Shahzad, M.; Ahmad, N.; Zamad, A.; Hussan, S.; Aslam, M.A.; Khan, A.R.; Asif, M.A.; Shahzadi, G.; Waseem, M. Performance Evaluation of Bus Rapid Transit System: A Comparative Analysis of Alternative Approaches for Energy Efficient Eco-Friendly Public Transport System. Energies 2020, 13, 1377. [Google Scholar] [CrossRef]
  33. Buehler, R.; Pucher, J. COVID-19 impacts on cycling, 2019–2020. Transp. Rev. 2021, 41, 393–400. [Google Scholar] [CrossRef]
  34. Combs, T.S.; Pardo, C.F. Shifting streets COVID-19 mobility data: Findings from a global dataset and a research agenda for transport planning and policy. Transp. Res. Interdiscip. Perspect. 2021, 9, 100322. [Google Scholar] [CrossRef]
  35. Woodward, C. Cities Want the Millions Who Biked during the Pandemic to Keep Pedaling. PBS News Hour. 2023. Available online: https://www.pbs.org/newshour/nation/cities-want-the-millions-who-biked-during-the-pandemic-to-keep-pedaling (accessed on 20 April 2024).
  36. Wilson, K. Cycling through COVID-19: Paris. StreetsBlog. USA. 2022. Available online: https://usa.streetsblog.org/2022/10/13/cycling-through-covid-19-paris-france (accessed on 29 January 2024).
  37. Buehler, R.; Pucher, J. Cycling through the COVID-19 pandemic to a more sustainable transport future: Evidence from case studies of 14 large bicycle-friendly cities in Europe and North America. Sustainability 2022, 14, 7293. [Google Scholar] [CrossRef]
  38. Rérat, P.; Haldimann, L.; Widmer, H. Cycling in the era of COVID-19: The effects of the pandemic and pop-up cycle lanes on cycling practices. Transp. Res. Interdiscip. Perspect. 2022, 15, 100677. [Google Scholar] [CrossRef] [PubMed]
  39. Doubleday, A.; Choe, Y.; Busch Isaksen, T.; Miles, S.; Errett, N.A. How did outdoor biking and walking change during COVID-19?: A case study of three US cities. PLoS ONE 2021, 16, e0245514. [Google Scholar] [CrossRef] [PubMed]
  40. Statistics Canada. (2017). Montréal [Census metropolitan area], Quebec and Canada Census Profile. 2016 Census. Statistics Canada Catalogue no. 98-316-X2016001. Available online: https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/prof/details/page.cfm?Lang=E<Geo1=CMACA<Code1=462<Geo2=PR<Code2=01<Data=Count<SearchText=montreal<SearchType=Begins<SearchPR=01<B1=All<TABID=1 (accessed on 3 November 2023).
  41. Copenhagenize. The Most Bicycle-Friendly Cities of 2019. 2019. Available online: https://copenhagenizeindex.eu/ (accessed on 12 February 2024).
  42. Vélo Québec. Cycling in Québec in 2020. 2020. Available online: https://www.velo.qc.ca/wp-content/uploads/2021/06/vq-edv2020-en.pdf (accessed on 7 April 2024).
  43. Autorité Régionale de Transport Métropolitain (ARTM). Enquête Origine-Destination—Montréal 2018. 2020. Available online: https://www.artm.quebec/planification/enqueteod/ (accessed on 20 May 2024).
  44. Ouelette-Vézina, H. Le REV Saint-Denis Plus Populaire Que Jamais. La Presse. 2023. Available online: https://www.lapresse.ca/actualites/grand-montreal/2023-12-15/pistes-cyclables/le-rev-saint-denis-plus-populaire-que-jamais.php#:~:text=Avec%20plus%20de%201%2C5,plus%20en%20plus%20d’adeptes (accessed on 6 March 2024).
  45. Ville de Montréal. Comptage des Vélos sur les Pistes Cyclables. Données Ouvertes. 2023. Available online: https://donnees.montreal.ca/dataset/velos-comptage (accessed on 7 December 2023).
  46. Moritz, S.; Bartz-Beielstein, T. ImputeTS: Time Series Missing Value Imputation in R. R J. 2017, 9, 207–218. [Google Scholar] [CrossRef]
  47. Young, M.; MacGregor, G.; Tanguay, A.G. Revisiting COVID-19’s impact on cycling: An examination of bicycle count data in Montreal. Transp. Find. 2024, 1–6. [Google Scholar] [CrossRef]
Figure 1. Montreal’s cycling network and sensors.
Figure 1. Montreal’s cycling network and sensors.
Sustainability 16 05992 g001
Figure 2. Methodology for uncovering the effect of the REV on nearby parallel bike routes.
Figure 2. Methodology for uncovering the effect of the REV on nearby parallel bike routes.
Sustainability 16 05992 g002
Figure 3. REV sensor-measured ridership comprising displaced cyclists and pent-up demand.
Figure 3. REV sensor-measured ridership comprising displaced cyclists and pent-up demand.
Sustainability 16 05992 g003
Table 1. Yearly cycling ridership per sensor.
Table 1. Yearly cycling ridership per sensor.
Ridership (×1000)
Sensor StreetGroup201820192020202120222023
Berri_OntarioControl95285541486610601165
CoteSainteCatherine_StuartControl418419407486486562
Maisonneuve_MarcilControl337347337329323302
Maisonneuve_PeelControl10591015603737966
NotreDame_FrontenacControl328269230252247280
Parc_DuluthControl593555336416583647
Rachel_HoteldeVilleControl710668659713716746
Rachel_PapineauControl103211421013104410131010
ReneLevesque_WolfeControl392413314375334325
University_MiltonControl688701272343528518
Viger_SaintUrbainControl13213944759196
Boyer_EverettParallel466449334390344299
Boyer_RosemontParallel711683447435432244
Brebeuf_RachelParallel741741383366345303
ChristopheColomb_LouvainParallel234223223247205205
SaintLaurent_BellechasseParallel13421399114010639821224
SaintUrbain_VilleneuveParallel409453288226227261
ClusterA_RachelREV 93211311298
ClusterB_CarrieresREV 112613211519
ClusterC_CastelnauREV 119759873
ClusterD_SauveREV 434442462
Table 2. Percent change in cycling ridership from the pre-COVID-19 average per sensor.
Table 2. Percent change in cycling ridership from the pre-COVID-19 average per sensor.
Pre-COVID-19 Average RidershipPercent Change from Pre-COVID-19 Average Ridership
NameGroup(2018–2019)2020202120222023
Berri_OntarioControl904−54.2−4.217.328.8
CoteSainteCatherine_StuartControl419−2.816.016.034.2
Maisonneuve_MarcilControl342−1.6−3.8−5.4−11.8
Maisonneuve_PeelControl1037−41.9−29.0−6.9
NotreDame_FrontenacControl299−23.0−15.8−17.5−6.4
Parc_DuluthControl574−41.5−27.51.612.8
Rachel_HoteldeVilleControl689−4.33.54.08.3
Rachel_PapineauControl1087−6.8−4.0−6.8−7.1
ReneLevesque_WolfeControl391−21.9−6.7−16.9−19.2
University_MiltonControl625−60.0−50.6−24.0−25.4
Viger_SaintUrbainControl135−67.5−44.1−32.6−28.7
Weighted averageControl −29.7−14.4−3.51.9
Boyer_EverettParallel (Cluster C)458−27.0−14.8−25.0−34.8
Boyer_RosemontParallel (Cluster B)697−35.9−37.6−38.0−65.0
Brebeuf_RachelParallel (Cluster A)741−48.3−50.7−53.4−59.1
ChristopheColomb_LouvainParallel (Cluster D)229−2.78.0−10.7−10.4
SaintLaurent_BellechasseParallel (Cluster B)1370−16.8−22.4−28.3−10.6
SaintUrbain_VilleneuveParallel (Cluster A)431−33.2−47.6−47.4−39.5
Weighted averageParallel −28.3−30.5−35.4−35.4
Table 3. Comparison of expected and actual drops in ridership along parallel bike routes per cluster.
Table 3. Comparison of expected and actual drops in ridership along parallel bike routes per cluster.
202120222023
Observed
Drop
Expected
Drop
Observed
Drop
Expected
Drop
Observed
Drop
Expected
Drop
Parallel sensors—Cluster A−785,606−213,156−803,837−37,663−778,26354,972
Parallel sensors—Cluster B−568,986−274,987−653,444−48,588−599,28270,918
Parallel sensors—Cluster C−67,586−60,877−114,060−10,757−158,92215,700
Parallel sensors—Cluster D18,687−30,419−24,153−5375−23,6007845
Table 4. The number and proportion of cyclists comprised within the displaced and pent-up demand categories per REV sensor over time.
Table 4. The number and proportion of cyclists comprised within the displaced and pent-up demand categories per REV sensor over time.
202120222023
Displaced (%)Pent-Up (%)Total (%)Displaced (%)Pent-Up (%)Total
(%)
Displaced (%)Pent-Up (%)Total
(%)
Cluster A572,450
(61)
359,820
(39)
932,270
(100)
766,174
(68)
364,953
(32)
1,131,127
(100)
833,235
(64)
464,616
(36)
1,297,851
(100)
Cluster B293,999
(26)
832,431
(74)
1,126,430
(100)
604,856
(46)
715,728
(54)
1,320,584
(100)
670,200
(44)
848,371
(56)
1,518,571
(100)
Cluster C6709
(6)
112,525
(94)
119,234
(100)
103,303
(14)
655,730
(86)
759,033
(100)
174,622
(20)
698,559
(80)
873,181
(100)
Cluster D0
(0)
433,923
(100)
433,923
(100)
18,778
(4)
423,508
(96)
442,286
(100)
31,445
(7)
430,216
(93)
461,661
(100)
Table 5. Percentage change in overall cycling volumes in Montreal compared to the average of 2018–2019.
Table 5. Percentage change in overall cycling volumes in Montreal compared to the average of 2018–2019.
2020202120222023
Change in cycling ridership measured by control sensors−28.7%−13.3%−2.4%3.4%
Change in cycling ridership measured by parallel sensors−28.3%−30.5%−35.4%−35.4%
Overall change in cycling ridership−28.5%−19.8%−14.8%−12.8%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Young, M.; MacGregor, G.; Tanguay, G.A. Analyzing the Impact of Montreal’s Réseau Express Vélo (REV) on Surrounding Bike Lanes’ Ridership and the COVID-19 Cycling Recovery. Sustainability 2024, 16, 5992. https://doi.org/10.3390/su16145992

AMA Style

Young M, MacGregor G, Tanguay GA. Analyzing the Impact of Montreal’s Réseau Express Vélo (REV) on Surrounding Bike Lanes’ Ridership and the COVID-19 Cycling Recovery. Sustainability. 2024; 16(14):5992. https://doi.org/10.3390/su16145992

Chicago/Turabian Style

Young, Mischa, Gavin MacGregor, and Georges A. Tanguay. 2024. "Analyzing the Impact of Montreal’s Réseau Express Vélo (REV) on Surrounding Bike Lanes’ Ridership and the COVID-19 Cycling Recovery" Sustainability 16, no. 14: 5992. https://doi.org/10.3390/su16145992

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop